Temperature measuring device and temperature measuring method

Abstract

To provide a temperature measuring device and a temperature measuring method with which it is possible to measure the temperature of a test sample at high speed.SOLUTION: A temperature measuring device 1 comprises: a light source 11 that generates laser light Lm for measurement that is collimated light having coherency; a high-speed camera 16 that detects the reflected light of the laser light Lm for measurement with which a test sample 31 is irradiated; a control unit 20 that acquires the vibration count and periodic change of a reflectance change waveform as reflectance information, on the basis of the intensity of the reflected light detected by the high-speed camera 16; and a database 50 that stores a plurality of pieces of temperature information including the reflectance information preliminarily acquired under different conditions and the temperature of the test sample 31. The control unit 20 selects, from the pieces of temperature information stored in the database 50, temperature information including reflectance information that is approximate to the reflectance information acquired on the basis of the intensity of reflected light, and calculates the temperature of the test sample 31 included in the selected temperature information as the measured temperature of the test sample 31.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a temperature measuring device and a temperature measuring method.

Background Art

[0002] Conventionally, a temperature measuring device has been developed that irradiates a sample to be measured with an interference light such as a laser beam and measures the temperature of the sample to be measured using the reflected light from the sample to be measured (for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In Patent Document 1, a pulsed laser beam having different wavelengths depending on the time from the start is irradiated to the measurement target, and based on the information on the difference in the intensity of the interference light due to the wavelength and the information on whether the intensity of the interference light is increasing or decreasing, it is determined whether the temperature of the measurement target is rising or falling. Further, the temperature is calculated based on the number of waves of the intensity change of the interference light.

[0005] However, in the method of Patent Document 1, since laser beams having different wavelengths are generated by a pulsed laser beam, complicated controls such as power control of the laser oscillator and control of the detection timing of the intensity of the interference light considering the wavelength change are required, and it is difficult to measure the temperature of the measurement target at high speed.

[0006] The present invention has been made in view of the above circumstances, and an object thereof is to provide a temperature measuring device and a temperature measuring method that can measure the temperature of a sample to be measured at high speed with a simple configuration.

Means for Solving the Problems

[0007] To achieve the above objective, the temperature measuring device according to the first aspect of this invention is: A light source that generates continuous light for measurement, which is collimated light with coherence properties, A detection device for detecting reflected light from the continuous measuring light irradiated onto the sample to be measured, A control unit that acquires the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform as reflectance information based on the intensity of the reflected light detected by the detection device, The system includes a database that stores the reflectance information and a plurality of temperature information, including the temperature of the sample to be measured, which have been acquired in advance under different conditions. The reflectance information includes first reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a first direction of an image of interference fringes caused by reflectance fluctuations, captured by the imaging device which is the detection device, and second reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. The control unit obtains from the temperature information stored in the database based on the intensity of the reflected light Reflectance information that approximates at least one of the reflectance information of the number of oscillations and period fluctuations of the reflectance fluctuation waveform. The temperature information including the selected temperature information is selected, and the temperature of the sample to be measured included in the selected temperature information is calculated as the measured temperature of the sample to be measured.

[0008] Furthermore, the temperature information stored in the database is Obtained from the calculation results of 3D heat conduction simulation, It would be acceptable to do so.

[0009] Furthermore, the temperature information stored in the database is The calculation results of the three-dimensional heat conduction simulation include the temperature distribution inside the sample being measured, It would be acceptable to do so.

[0011] Furthermore, the control unit, Based on the periodic fluctuations obtained from the intensity of the reflected light detected by the detection device and the periodic fluctuations included in the temperature information stored in the database, the temperature information that minimizes the sum of the squares of the differences between each period is selected as the approximate temperature information including the reflectance information. It would be acceptable to do so.

[0012] Further, the temperature measurement device according to the second aspect of the present invention is a light source that generates measurement continuous light, which is collimated light having coherence, a detection device that detects the reflected light of the measurement continuous light irradiated on the measurement sample, a control unit that acquires the vibration frequency and periodic variation of the reflectance fluctuation waveform as reflectance information based on the intensity of the reflected light detected by the detection device, and a storage unit that stores an estimation model generated by machine learning so as to estimate the temperature of the measurement sample from the reflectance information using, as teacher data, a plurality of temperature information including the reflectance information acquired in advance under different conditions Reflectance information of at least one of the oscillation frequency and period variation of the reflectance fluctuation waveform and the temperature of the measurement sample, The aforementioned reflectance information is, The detection device includes first reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a first direction of the image of interference fringes caused by reflectance fluctuations captured by the imaging device, and second reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. The control unit inputs the acquired Reflectance information of at least one of the oscillation frequency and period variation of the reflectance fluctuation waveform into the estimation model and estimates the temperature of the measurement sample based on the output result of the estimation model.

[0013] Further, in the temperature measurement method according to the third aspect of the present invention, the measurement sample is irradiated with measurement continuous light, which is collimated light having coherence, the reflected light of the measurement continuous light is detected, based on the intensity of the detected reflected light, The image includes, captured by the imaging device, an image of interference fringes caused by reflectance fluctuations, and includes first reflectance information including the number of oscillations and period fluctuations of the reflectance fluctuation waveform acquired in a first direction, and second reflectance information including the number of oscillations and period fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. the reflectance information is acquired, from a plurality of temperature information including the reflectance information acquired in advance under different conditions and stored in a database and the temperature of the measurement sample, the temperature information including Reflectance information that approximates at least one of the reflectance information of the number of oscillations and period fluctuations of the reflectance fluctuation waveform. acquired based on the intensity of the reflected light is selected, and the temperature of the measurement sample included in the selected temperature information is calculated as the measurement temperature of the measurement sample.

Advantages of the Invention

[0014] According to the temperature measurement device and temperature measurement method of the present invention, the temperature of the measured sample is measured based on the reflectance information consisting of the number of vibrations and periodic fluctuations in the reflectance fluctuations caused by the temperature change of the measured sample, and a database or estimation model related to the reflectance information and temperature information including the previously acquired reflectance information and the temperature of the measured sample. Therefore, it is possible to measure the temperature at high speed with a simple configuration.

Brief Description of the Drawings

[0015] [Figure 1] It is a diagram showing a schematic configuration of the temperature measurement device according to Embodiment 1 of the present invention. [Figure 2] It is a functional block diagram of the control unit according to Embodiment 1. [Figure 3] It is a flowchart showing the flow of temperature measurement according to Embodiment 1. [Figure 4] It is a conceptual diagram showing the optical path of the measurement laser light in the measured sample. [Figure 5] It is a graph showing an example of the reflectance fluctuation waveform due to the temperature change of the measured sample. [Figure 6] It is a diagram showing an example of temperature information according to Embodiment 1. [Figure 7] It is a flowchart showing the flow of temperature measurement according to Embodiment 2. [Figure 8] It is a diagram showing an example of the calculation result of the three-dimensional heat conduction simulation. [Figure 9] It is a diagram showing an example of the interference fringes created from the calculation result of the three-dimensional heat conduction simulation. [Figure 10] It is a diagram showing an example of the photographed interference fringe image. [Figure 11] (A) is a diagram obtained by synthesizing the interference fringe image of FIG. 10, and (B) is a diagram obtained by performing image processing on the diagram of (A). [Figure 12] It is a functional block diagram of the control unit according to Embodiment 3.

Modes for Carrying Out the Invention

[0016] (Embodiment 1) The following describes a temperature measuring device 1 according to an embodiment of the present invention, with reference to the figures. In this embodiment, the case of measuring the temperature of a SiC (silicon carbide) wafer, which is the sample to be measured, when heat treatment is performed on the SiC wafer using an atmospheric pressure plasma device will be described as an example.

[0017] As shown in Figure 1, the temperature measuring device 1 according to this embodiment includes a light source 11, a beam splitter 14, a bandpass filter 15, a high-speed camera 16, a control unit 20, and a database 50. The light source 11 also includes a laser oscillator 11a, a reflector 11b, and a beam expander 11c. The sample to be measured, a SiC wafer, is a flat plate-shaped sample having a front and back surface that are parallel to each other, and its surface is heated by a micro-plasma jet (μ-TPJ), which is a high-density atmospheric pressure thermal plasma jet (TPJ) with a plasma size of less than a millimeter. The sample to be measured 31 and the temperature measuring device 1 move in parallel while maintaining a constant distance from the plasma device 41 that generates the μ-TPJ. The temperature measuring device 1 then measures the temperature of the sample to be measured 31 based on the change in the reflectance of the laser light detected on the back surface of the sample to be measured 31.

[0018] The light source 11 is an optical system that generates continuous measurement light, which is collimated light with coherence used for measurement. The light source 11 according to this embodiment includes a laser oscillator 11a that generates continuous measurement light (hereinafter referred to as measurement laser light Lm). The measurement laser light Lm is not particularly limited, but one with a wavelength that can penetrate the sample to be measured 31 is selected. For example, when alkali-free glass, quartz glass, silicon carbide, gallium nitride, etc. are used as the sample to be measured 31, a visible light laser can be used as the measurement laser light Lm. The measurement laser light Lm according to this embodiment is a helium-neon (He-Ne) laser.

[0019] The reflector 11b is positioned in the optical path between the laser oscillator 11a and the beam expander 11c, and directs the measurement laser light Lm emitted from the laser oscillator 11a into the beam expander 11c.

[0020] The beam expander 11c receives the measurement laser beam Lm emitted from the laser oscillator 11a, expands it to the temperature measurement range, and emits the expanded collimated measurement laser beam Lm. The measurement range is the measurement area for interference fringes caused by reflectance fluctuations based on the temperature change of the sample 31 under measurement. The expansion rate of the measurement laser beam Lm expanded by the beam expander 11c is not particularly limited and should be set within a range where the power during expansion does not become too small, based on the power of the measurement laser beam Lm generated by the light source 11. Furthermore, by making the measurement area wider than the heat input range, the observation of interference fringes becomes easier. In this embodiment, the beam expander 11c expands the measurement laser beam Lm, which is incident with a diameter of 0.6 mm, to a diameter of 1.0 mm.

[0021] The beam splitter 14 reflects the measurement laser light Lm, which has been expanded by the beam expander 11c, and irradiates the measurement area of ​​the sample 31 perpendicularly. The beam splitter 14 also transmits the measurement laser light Lm reflected from the sample 31 and directs it into the high-speed camera 16. The positional relationship between the beam expander 11c and the high-speed camera 16 with respect to the beam splitter 14, that is, which is placed on the reflecting side and which is placed on the transmitting side, is not particularly limited.

[0022] The bandpass filter 15 is placed in the optical path between the sample 31 under measurement and the high-speed camera 16 to eliminate stray light, which is light other than the measurement laser light Lm. By passing light through the bandpass filter 15, a clearer image of the reflected light from the sample 31 under measurement can be obtained.

[0023] The bandpass filter 15 according to this embodiment is placed between the sample to be measured 31 and the beam splitter 14, as shown in Figure 1, but is not limited to this. For example, in order to reduce the loss of reflected light, the bandpass filter 15 may be placed between the beam splitter 14 and the high-speed camera 16.

[0024] The high-speed camera 16 is a detection device that detects reflected light from a measurement laser beam Lm, and is an imaging device comprising a camera body 16a having a photoreceiving element and an objective lens 16b. The high-speed camera 16 uses a sensor sensitive to the wavelength of the measurement laser beam Lm, and magnifies the reflected light from the sample to be measured 31 with the objective lens 16b, which is then captured by the camera body 16a. The frame rate of the high-speed camera 16 is set to have an appropriate time resolution for the time change of the temperature distribution of the sample to be measured 31 being observed, for example, it is set to between 3000fps and 10000fps.

[0025] The control unit 20 is, for example, a computer device and includes a control unit 21, a storage unit 22, a display unit 23, and an input unit 24, as shown in the functional block diagram of Figure 2.

[0026] The control unit 21 consists of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), a crystal oscillator, etc., and controls the operation of the temperature measuring device 1 and calculates the temperature of the sample to be measured 31 based on the interference fringe image captured by the high-speed camera 16.

[0027] The control unit 21 loads various operation programs and data stored in the ROM, memory unit 22, etc., into RAM and operates the CPU, thereby realizing the various functions of the control unit 21 shown in Figure 2. As a result, the control unit 21 operates as a light source control unit 211, an image processing unit 212, and an arithmetic unit 213.

[0028] The light source control unit 211 controls the light source 11 to generate and stop the measurement laser beam Lm. Since the measurement laser beam Lm in this embodiment is continuous light, complex control such as power supply control for generating pulsed laser light is unnecessary.

[0029] The image processing unit 212 acquires reflectance information from image data captured by the high-speed camera 16. The reflectance information is used to calculate the temperature of the sample 31 to be measured, and the reflectance information in this embodiment includes the number of oscillations and periodic fluctuations in the reflectance fluctuation waveform of the laser light.

[0030] The calculation unit 213 calculates the temperature of the sample to be measured 31 based on the reflectance information acquired by the image processing unit 212 and the temperature information, including the reflectance information, that is pre-stored in the database 50.

[0031] The memory unit 22 is a non-volatile memory such as a hard disk or flash memory, and stores images captured by the high-speed camera 16, data such as the thermo-optic coefficient representing the temperature dependence of the refractive index of the sample under test 31, and calculation algorithms for calculating the temperature of the sample under test 31 from the temperature information.

[0032] The display unit 23 is a display device provided in the control unit 20, which is a computer device, and is, for example, a liquid crystal panel. The display unit 23 displays images of interference fringes caused by reflectance fluctuations captured by the high-speed camera 16, the temperature of the sample under measurement 31 calculated by the calculation unit 213, and so on.

[0033] The input unit 24 is an input device for inputting instructions to start and end measurements in the temperature measuring device 1, changes to various evaluation conditions, etc. The input unit 24 is a keyboard, touch panel, mouse, etc., provided on the control unit 20.

[0034] The database 50 is a non-volatile memory such as a hard disk or flash memory connected to the control unit 20. The database 50 stores temperature information related to various conditions that have been acquired in advance, that is, information regarding the temperature when the sample to be measured 31 is heated under various conditions. Details of the content of the temperature information and the acquisition method will be described later.

[0035] Next, the temperature measurement method using the temperature measuring device 1 described above will be explained with reference to the flowchart in Figure 3. In this embodiment, heat is applied to the surface of the sample 31 by a microplasma jet (μ-TPJ) generated by the plasma device 41 for heat treatment of the sample 31. The temperature measuring device 1 then irradiates the back surface of the sample 31 with a measuring laser beam Lm to measure the temperature of the sample 31, particularly the highest temperature reached by the heat treatment.

[0036] In this embodiment, a μ-TPJ is used as the thermal input, but it is not limited to this. Pulsed light such as a pulsed laser or flash lamp, a CW (Continuous Wave) laser, or Joule heating by a heating element placed on the sample surface may also be used.

[0037] First, as part of the database creation process, a heat conduction simulation is performed using a computer to calculate the temperature distribution related to various conditions such as heat input to the sample and the thermal properties (density, specific heat, thermal conductivity) of the sample to be measured 31 (step S11). The heat conduction simulation and other processes related to the database creation process may be performed by the control unit 20 of the temperature measuring device 1, or by using another computer device. In this embodiment, the database is created using an external computer device.

[0038] Here, we will explain the reflectance of the reflected light of the measurement laser beam Lm. The measurement laser beam Lm is reflected from the back surface of the sample 31 and also passes into the interior of the sample 31. A portion of the measurement laser beam Lm that has passed into the interior of the sample 31 is reflected from the front surface of the sample 31 and then reflected again from the back surface of the sample 31.

[0039] Since the refractive index of the sample 31 under test is temperature-dependent, the interference state of the multiple reflections of the measurement laser light Lm differs depending on the temperature of each part of the sample 31, and the intensity of the reflected light changes. More specifically, as shown in the conceptual diagram of Figure 4, the measurement laser light Lm undergoes multiple reflections on the front and back surfaces of the sample 31, and the reflected light interferes with each other. Therefore, the reflectance R of the measurement laser light Lm reflected to the back surface of the sample 31 is determined by the superposition of the phases of the multiple reflections of the measurement laser light Lm. Thus, the reflectance R is determined based on the refractive index, thickness, and wavelength of light of the sample 31 under test. In the case of a uniform temperature distribution without a temperature gradient, the reflectance R is the same in the plane direction and no fringes are produced. When there is a temperature distribution in the sample 31 under test, interference fringes are observed due to the distribution of reflectance R. That is, the change in reflectance obtained by measurement represents a temperature change, and the image of the interference fringes can be considered as an image representing the contour lines of the temperature distribution.

[0040] The temperature of the sample 31 under test, corresponding to the peaks and troughs (maximum and minimum values) of the reflectance fluctuation waveform appearing as interference fringes, can be obtained in advance by a heat conduction simulation based on conditions such as the physical properties and moving speed of the sample 31, the amount of heat input via μ-TPJ, and the wavelength of the measurement laser light Lm. In this embodiment, these conditions are varied and a 3D heat conduction simulation is performed using a computer to obtain the time change of the temperature at the measurement point under each condition. Then, the reflectance fluctuation is calculated from the obtained temperature change data, and the number of oscillations and period fluctuations (reflectance information) in the reflectance fluctuation waveform are obtained. The period related to the period fluctuation is not particularly limited, and the time interval between consecutive maximum values, the time interval between consecutive minimum values, the time interval of inflection points, etc., can be used. In this embodiment, the time interval between consecutive maximum and minimum values ​​is used as the period.

[0041] Figure 5 shows an example of reflectance fluctuation at the measurement site. As shown in Figure 5, the period of reflectance fluctuation during temperature rise decreases from period P1 at the start of measurement to period P2, and increases from period P2 to period P3, which represents the highest temperature reached. In this way, by observing the number of oscillations and period fluctuations in reflectance fluctuation, it is possible to understand the state of temperature change, such as whether the temperature is rising or falling, and the degree of temperature change, thereby enabling the measurement of the highest temperature reached by the sample 31 under test.

[0042] As shown in Figure 6, in this embodiment, based on the relationship between the temperature and reflectance of the sample 31 to be measured obtained from the calculation results of the 3D heat conduction simulation, a set of datasets including the number of oscillations, periodic fluctuations, and maximum temperature reached of the reflectance fluctuation waveform related to the measurement surface of the sample 31 to be measured is stored in the database 50 as temperature information (step S12). Furthermore, the number of oscillations and periodic fluctuations related to the reflectance information shall include data at least when the temperature rises.

[0043] Next, as part of the measurement process, heat is applied by the plasma device 41, and the measurement begins. Once the measurement starts, the control unit 21 controls the high-speed camera 16 to start taking images (step S13). The light source control unit 211 also controls the laser oscillator 11a of the light source 11 to emit the measurement laser beam Lm. As shown in Figure 1, the measurement laser beam Lm is incident on the beam expander 11c via the reflector 11b. The measurement laser beam Lm, which is collimated light expanded by the beam expander 11c, is reflected by the beam splitter 14 and irradiated onto the sample to be measured 31 via the bandpass filter 15.

[0044] The measurement laser beam Lm is reflected from the back surface of the sample 31 and also transmitted into the interior of the sample 31. A portion of the measurement laser beam Lm that has been transmitted into the interior of the sample 31 is reflected from the front surface of the sample 31 and then reflected again from the back surface of the sample 31. As a result, the measurement laser beam Lm, which has undergone multiple reflections from the front and back surfaces of the sample 31, passes through the bandpass filter 15 and the beam splitter 14 and is incident on the high-speed camera 16.

[0045] As described above, the refractive index of the sample 31 under test is temperature-dependent. Therefore, the interference state of the multiple reflections of the measurement laser beam Lm differs depending on the temperature of each part of the sample 31 under test, causing the intensity (reflectance) of the reflected light to fluctuate. Due to the fluctuation in reflectance, interference fringes are observed on the measurement surface of the sample 31 under test. In other words, the high-speed camera 16 according to this embodiment detects the reflected light of the measurement laser beam Lm by photographing the interference fringes.

[0046] The image processing unit 212 acquires images captured by the high-speed camera 16 and obtains reflectance information, namely the number of vibrations and periodic fluctuations, from the peaks and valleys of reflectance fluctuations that appear as interference fringes (step S14). The image processing unit 212 sequentially acquires images captured by the high-speed camera 16 and obtains the number of vibrations and periodic fluctuations until the heat treatment is completed, and stores them in the storage unit 22.

[0047] The calculation unit 213 searches the database 50 for temperature information and selects the temperature information containing the reflectance information that most closely matches the reflectance information obtained in step S14 as approximate data (step S15). The method for selecting approximate data is not particularly limited, but for example, a method can be used to select the temperature information that minimizes the sum of the squares of the differences between each period, based on the periodic fluctuations related to the measurement data and the periodic fluctuations of the temperature information stored in the database 50.

[0048] The calculation unit 213 displays the highest temperature reached, which is included in the temperature information selected as approximate data, on the display unit 23 as the measured temperature of the sample 31, and also stores it in the storage unit 22 (step S16).

[0049] As described above, according to the temperature measuring device and temperature measuring method of this embodiment, The reflectance information, including the number of oscillations and periodic fluctuations in the waveform of the reflectance fluctuation caused by the temperature change of the sample 31 under test, is compared with the reflectance information in the temperature information stored in the database 50 to select approximate data. Then, the temperature of the sample under test is measured based on the temperature information selected as approximate data. Therefore, it is possible to measure temperature with a simpler configuration compared to using laser light of multiple wavelengths. In addition, by using the period of the reflectance fluctuation as a feature quantity, approximate data can be selected based only on the reflectance information, making it possible to measure temperature quickly with a small amount of computation.

[0050] In this embodiment, the calculation unit 213 selects approximate data based on reflectance information from the start to the end of the heat treatment, but it is not limited to this, and may also select approximate data based only on data during temperature rise. This allows for high-speed temperature measurement with less computation, for example, when only the highest temperature reached needs to be measured.

[0051] Furthermore, in the temperature measuring device 1 according to this embodiment, a high-speed camera 16 is used as a detection device to capture interference fringes of reflected light and acquire reflectance information, but the device is not limited to this. For example, instead of a camera, a light intensity measuring device such as a photodiode may be used as a detection device to acquire time-series data of the reflectance of the measurement location. In this case, the calculation unit 213 can obtain reflectance information from the acquired time-series data of reflectance and compare it with the reflectance information contained in the temperature information stored in the database 50 to calculate the temperature change of the measurement location, the highest temperature reached, etc. As a result, the temperature of the sample to be measured 31 can be measured without analyzing the image captured by the high-speed camera 16, so that temperature measurement can be performed at high speed with a simpler configuration.

[0052] Furthermore, while the temperature measuring device 1 according to this embodiment is used to measure the temperature related to the heat treatment of a SiC wafer, it is not limited to this. The object of temperature measurement can be anything that can detect the intensity of reflected light due to temperature changes.

[0053] Furthermore, while the light source 11 in this embodiment expands the measurement laser light Lm generated by the laser oscillator 11a using the beam expander 11c, it is not limited to this configuration, and any configuration that generates collimated light corresponding to the size of the measurement area is acceptable. For example, a laser oscillator 11a that generates collimated light corresponding to the measurement area may be used as the light source 11.

[0054] Furthermore, in this embodiment, the beam splitter 14 is placed between the bandpass filter 15 and the objective lens 16b of the high-speed camera 16, but this is not the only option. The beam splitter 14 can be placed between the sample to be measured 31 and the objective lens 16b.

[0055] In this embodiment, the temperature measuring device 1 and the sample to be measured 31 are moved relative to the plasma device 41, but this is not limited to this. For example, the temperature measuring device 1 and the sample to be measured 31 may be fixed, and the temperature measurement may be performed while moving the heat input device such as the plasma device 41. Alternatively, the temperature measurement may be performed with the temperature measuring device 1, the sample to be measured 31, and the plasma device 41 fixed. Furthermore, the temperature distribution in a steady heating state may be measured while moving the temperature measuring device 1 with the sample to be measured 31 and the plasma device 41 fixed.

[0056] Furthermore, while the temperature information stored in the database 50 according to this embodiment is assumed to be data acquired in advance based on the calculation results of a three-dimensional heat conduction simulation, it is not limited to this, and temperature information based on actual measurement results may also be used.

[0057] (Embodiment 2) In the above embodiment 1, the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform are acquired as reflectance information from images captured by the high-speed camera 16, and the temperature of the sample to be measured 31 is measured by comparing it with the reflectance information included in the temperature information obtained in advance from the calculation results of a heat conduction simulation based on various conditions. However, other information can also be used as reflectance information and temperature information. In this embodiment, a temperature measuring device and temperature measuring method will be described that use multiple reflectance information acquired from interference fringe images created based on time-series images of the measurement location as reflectance information.

[0058] In this embodiment, the temperature information related to the measurement data and the data stored in the database 50 differs from that of Embodiment 1. Furthermore, the configuration of the optical system that irradiates the sample 31 with the measurement laser light Lm and captures interference fringes is the same as in Embodiment 1, so the same reference numerals are used and a detailed explanation is omitted.

[0059] The temperature measurement method according to this embodiment will be described below with reference to the flowchart in Figure 7.

[0060] First, as part of the database creation process, a 3D heat conduction simulation using a computer is performed, similar to Embodiment 1, to calculate the temperature distribution under various conditions (step S21). Figure 8 shows an example of the calculation results of the 3D heat conduction simulation, illustrating how the temperature distribution of the sample 31 under measurement changes over time.

[0061] Based on the temperature distribution obtained from the 3D heat conduction simulation and the material properties of the sample 31 under test, an interference fringe image representing the reflectance variation is created (step S22).

[0062] Reflectance information is obtained based on the created interference fringe image. Specifically, as shown in Figure 9, reflectance information (hereinafter referred to as the first reflectance information) is calculated in the direction of movement of the sample under test 31 (the x-axis direction in Figure 9; hereinafter also referred to as the first direction) centered on the point with the highest temperature in the created interference fringe image. Reflectance information (hereinafter referred to as the second reflectance information) is also calculated in the direction passing through the point with the highest temperature and perpendicular to the first direction (the y-axis direction in Figure 9; hereinafter also referred to as the second direction). Then, the reflectance information including the calculated first and second reflectance information, and temperature information including the results of a 3D heat conduction simulation including the highest temperature reached, are stored in the database 50 (step S23).

[0063] Next, as part of the measurement process, heating by the plasma device 41 is started, similar to Embodiment 1, and the sample to be measured 31 and the optical system are moved, and the measurement is started. Once the measurement is started, the control unit 21 controls the high-speed camera 16 to acquire an image of the interference fringes appearing on the measurement surface of the sample to be measured 31 at a predetermined sampling frequency and stores it in the storage unit 22 (step S24).

[0064] The image processing unit 212 performs image processing on the image captured in step S24 to enhance interference fringes (step S25). Specifically, the image processing unit 212 performs background removal and contrast adjustment processing on the captured image (Figure 10).

[0065] The image processing unit 212 performs a synthesis process to combine the images of each frame adjusted in step S25 (Figure 11(A)). Furthermore, in order to appropriately acquire interference fringe information, the image processing unit 212 performs noise reduction on the synthesized image and then performs binarization (Figure 11(B)). From the denoised synthesized image, the image processing unit 212 acquires reflectance information contained in the temperature information, i.e., the number of oscillations and period fluctuations of the reflectance fluctuation waveform (step S26).

[0066] The acquisition of reflectance information is first performed in a first direction, which is the center of the measurement area, passes through the heating area by the plasma device 41, and is the direction of movement of the sample to be measured 31 (the x-axis direction in Figure 11(B)). The image processing unit 212 estimates the part where the highest temperature is reached (hereinafter also referred to as the temperature center point) from the first reflectance information acquired in the first direction, and acquires second reflectance information in a second direction (the y-axis direction in Figure 11(B)) that passes through the temperature center point and is perpendicular to the first direction.

[0067] The calculation unit 213 searches the database 50 for temperature information with respect to the reflectance information, including the first and second reflectance information acquired by the image processing unit 212, and selects the temperature information containing the most similar reflectance information as approximate data through pattern matching (step S27). More specifically, it calculates the similarity of the number of oscillations and period fluctuations of the reflectance fluctuation waveform, which is the reflectance information, and selects the data that best approximates it. As a method for calculating the similarity, for example, the calculation unit 213 calculates the error by summing the squares of each period acquired by measurement and each period related to the temperature information stored in the database 50 in the +x, -x, +y, and -y directions from the temperature center point. Then, the calculation unit 213 selects the temperature information with the smallest error as the temperature information with the highest similarity, i.e., approximate data.

[0068] The control unit 21 reads the data of the highest reached temperature and temperature distribution included in the temperature information, which is the selected approximate data, from the database 50 and displays it on the display unit 23 (step S28).

[0069] As described above, the temperature measuring device and temperature measuring method according to this embodiment select approximate data using multiple reflectance information obtained from interference fringe images created based on time-series images of the measurement location, so that the temperature of the sample to be measured 31 can be measured quickly and with greater accuracy.

[0070] In this embodiment, the temperature information stored in the database 50 includes the calculation results of a three-dimensional heat conduction simulation. Specifically, the temperature distribution of the sample 31 under test for each simulation and the corresponding reflectance information are stored in the database 50. This makes it possible to easily estimate the temperature distribution inside the sample 31 under test (Figure 8) based on the matching of the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform on the measurement surface of the sample 31 under test.

[0071] Furthermore, in this embodiment, the first direction is the direction of movement of the sample 31 under measurement, and the second direction is the direction perpendicular to the first direction, but this is not limited to this. The first and second directions should be selected in directions that make it easy to grasp characteristics such as the number and period of interference fringes.

[0072] In the embodiments described above, a database 50 is provided separately from the control unit 20, but this is not the only option. For example, the storage unit 22 of the control unit 20 may also serve as the database 50 and store temperature information. Alternatively, the database 50 may be connected to the control unit 20 via a network. This allows multiple temperature measuring devices 1 to share the database 50.

[0073] (Embodiment 3) In Embodiment 2, approximate data was selected by pattern matching based on reflectance information to measure the temperature of the sample 31 under test. However, it is also possible to measure the temperature using a pre-trained model obtained through machine learning. This embodiment differs from Embodiment 2 in that the calculation unit 213 of the control unit 21 measures the temperature of the sample 31 under test from the acquired reflectance information using a pre-trained model TM obtained through machine learning. It also differs from Embodiment 2 in that the temperature measuring device 1 does not have a database 50. The configuration of the temperature measuring device 1 is the same as in Embodiment 2, so the same reference numerals are used and detailed explanations are omitted.

[0074] As shown in Figure 12, the memory unit 22, similar to Embodiment 2, includes a trained model TM that has been trained using a three-dimensional heat conduction simulation to learn reflectance fluctuations related to temperature distributions calculated under various conditions as training data. The trained model TM is an estimation model generated by machine learning to estimate the temperature of the sample under test 31 from reflectance information such as the period of reflectance fluctuations appearing in the interference fringe image. For example, machine learning is performed using reflectance information calculated based on the temperature distribution, which is the result of a three-dimensional heat conduction simulation related to various conditions such as heat input to the sample and the thermal properties of the sample under test 31 (density, specific heat, thermal conductivity), as training data.

[0075] The machine learning algorithm is not particularly limited, but for example, it is a machine learning algorithm using a neural network. In this embodiment, as described above, a trained model TM is generated in advance using the period of reflectance fluctuations that appear as interference fringe images as a feature. The generated trained model TM is then stored in the memory unit 22.

[0076] In this embodiment, the database creation process (steps S21 to S23) shown in the flowchart of Figure 7 is replaced by the creation of the trained model TM described above. Furthermore, the temperature measuring device 1 performs temperature measurement using the trained model TM instead of steps S27 and S28 in the flowchart of Figure 7. Specifically, the calculation unit 213 reads the trained model TM stored in the storage unit 22 and inputs the reflectance information acquired in step S26 into the trained model TM. The resulting output, i.e., the estimated temperature of the sample to be measured 31, is then displayed on the display unit 23 as the measured temperature of the sample to be measured 31.

[0077] As described above, the temperature measuring device 1 according to this embodiment is equipped with a trained model TM based on previously acquired reflectance information, and the calculation unit 213 measures the temperature of the sample to be measured 31 using the trained model TM. Therefore, unlike pattern matching according to Embodiment 2, there is no need to sequentially compare the reflectance information of the interference fringe image with the reflectance information stored in the database 50, so the temperature of the sample to be measured 31 can be easily measured.

[0078] Furthermore, the number of temperature data including reflectance information used to create the trained model TM can be significantly less than the number of data in the database used for pattern matching based on reflectance information according to Embodiment 2. For example, a trained model TM can be created with about one-tenth the number of data related to various conditions for pattern matching. Therefore, it is possible to reduce the time and cost for measurement preparation, such as the database creation process.

[0079] In this embodiment, the training data is assumed to be data acquired in advance based on the calculation results of a 3D heat conduction simulation, but it is not limited to this, and temperature information based on actual measurement results may also be used. [Industrial applicability]

[0080] The present invention is suitable for a temperature measuring device that measures the temperature distribution of a sample under test at high speed. In particular, it is suitable for a temperature measuring device that measures temperature changes during the heat treatment of semiconductor wafers, where temperature changes occur rapidly in a short time. [Explanation of Symbols]

[0081] 1 Temperature measuring device, 11 Light source, 11a Laser oscillator, 11b Reflector, 11c Beam expander, 14 Beam splitter, 15 Bandpass filter, 16 High-speed camera, 16a Camera body, 16b Objective lens, 20 Control unit, 21 Control unit, 211 Light source control unit, 212 Image processing unit, 213 Calculation unit, 22 Storage unit, 23 Display unit, 24 Input unit, 31 Sample to be measured, 41 Plasma device, 50 Database, Lm Laser light for measurement, TM Trained model

Claims

1. A light source that generates continuous light for measurement, which is collimated light with coherence properties, A detection device for detecting reflected light from the continuous measuring light irradiated onto the sample to be measured, A control unit that acquires the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform as reflectance information based on the intensity of the reflected light detected by the detection device, The system includes a database that stores the reflectance information and a plurality of temperature information, including the temperature of the sample to be measured, which have been acquired in advance under different conditions. The reflectance information includes first reflectance information including the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a first direction of an image of interference fringes caused by reflectance fluctuations, which is captured by the imaging device that is the detection device, and second reflectance information including the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. The control unit selects from the temperature information stored in the database the temperature information that includes reflectance information that approximates at least one of the reflectance information of the number of oscillations and period fluctuations of the reflectance fluctuation waveform obtained based on the intensity of the reflected light, and calculates the temperature of the sample to be measured included in the selected temperature information as the measured temperature of the sample to be measured. A temperature measuring device characterized by the following features.

2. The temperature information stored in the database is Obtained from the calculation results of a 3D heat conduction simulation, The temperature measuring device according to feature 1.

3. The temperature information stored in the database is The calculation results of the three-dimensional heat conduction simulation include the temperature distribution inside the sample being measured, The temperature measuring device according to feature 2.

4. The control unit, Based on the periodic fluctuations obtained from the intensity of the reflected light detected by the detection device and the periodic fluctuations included in the temperature information stored in the database, the temperature information that minimizes the sum of the squares of the differences between each period is selected as the approximate temperature information including the reflectance information. A temperature measuring device according to any one of claims 1 to 3.

5. A light source that generates continuous light for measurement, which is collimated light with coherence properties, A detection device for detecting reflected light from the continuous measuring light irradiated onto the sample to be measured, A control unit that acquires the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform as reflectance information based on the intensity of the reflected light detected by the detection device, The system includes a storage unit that stores an estimation model generated by machine learning, which uses reflectance information, including at least one of the number of oscillations and period fluctuations of reflectance fluctuation waveforms acquired in advance under different conditions, and a plurality of temperature information including the temperature of the sample to be measured, as training data to estimate the temperature of the sample to be measured from the reflectance information. The aforementioned reflectance information is, The detection device includes first reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a first direction of the image of interference fringes caused by reflectance fluctuations captured by the imaging device, and second reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. The control unit, The reflectance information of at least one of the oscillation frequency and period fluctuations of the acquired reflectance fluctuation waveform is input to the estimation model, and the temperature of the sample under measurement is estimated based on the output result of the estimation model. A temperature measuring device characterized by the following features.

6. The sample to be measured is irradiated with continuous measurement light, which is collimated light with coherence properties. The reflected light of the aforementioned continuous measurement light is detected, Based on the intensity of the detected reflected light, reflectance information is obtained that includes first reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a first direction of the image of interference fringes caused by reflectance fluctuations captured by the imaging device, and second reflectance information, which includes the number of oscillations and periodic fluctuations of the reflectance fluctuation waveform acquired in a second direction orthogonal to the first direction. From a plurality of temperature information, including the reflectance information and the temperature of the sample to be measured, which have been acquired in advance under different conditions and stored in a database, select the temperature information that includes reflectance information that approximates at least one of the reflectance information of the number of oscillations and period fluctuations of the reflectance fluctuation waveform acquired based on the intensity of the reflected light. The temperature of the sample to be measured, included in the selected temperature information, is calculated as the measured temperature of the sample to be measured. A method for measuring temperature characterized by the following features.

Images